The risk of GVHD may be less when cord blood (containing naive T cells) rather than adult BM (containing memory T cells) is infused. Male recipients of donor BM from multiparous females and recipients exposed to herpes virus pre-BMT have a higher incidence of GVHD despite post-BMT immunosuppression. These data suggest that memory T cells differ from naive T cells in their capacity or requirements for mediating GVHD. This project is directed toward investigating the potential biological differences between naive and memory T cells in responses relevant to BMT using new models of adoptive T cell transfer to analyze alloresponses and peptide-specific responses. For characterizing the requirements for alloresponses, GVHD studies will be performed in primary BMT recipients of allogeneic lymph node-T cells and secondary recipients of allosensitized memory T cells which have been generated by parking of GVHD effector cells prior to isolation and infusion into secondary BMT recipients. Primary or secondary recipients will be treated with monoclonal antibodies (mAbs) blocking one or more costimulatory/adhesogenic (B7-CD28; LFA-1:ICAM; CD2:CD48;OX40:OX40L) pathways, and alloreactive donor T cells, isolated by thoracic duct cannulation, will be analyzed to quantify the degree of expansion, activation/memory cell conversion, and function. We will use the D0.11 adoptive transfer model to follow responses of naive or memory D0.11 T cells to Oval peptide in the presence of mAbs known to inhibit GVHD responses, and analyze D0.11 T cell responses by quantifying D0.11 T cell expansion, activation and function. To determine whether costimulatory blockade has selectively influenced T cell responses to Ova and not altered T cell function to other antigens, we will perform studies in which adoptive transfer recipients will be immunized with a viral peptide antigen prior to adoptive transfer, the follow both viral and Ova peptide responses after immunizing these mice with Ova in the presence of costimulatory blockade. Pathways deemed critical for conversion of naive to memory T cell responses and prevention of GVHD lethality and peptide responsiveness will be further studied using D0.11 mice lacking these constimulatory ligands/receptors. These studies will establish the hierarchy of requirements for the pathways examined and will enable the design of GVHD preventive approaches that take into account the hierarchial pathways important for alloresponses.